[1] genomics of DNA repair

We aim to understand mechanisms of maintaining genome integrity in human cells via statistical analyses of mutation patterns.

Protecting what matters. In human somatic cells mutation rates are very uneven across the genome, with consequences on risk of cancer and, possibly, other age-related pathologies. We think somatic mutations are unevenly distributed due to differential activity of DNA mismatch repair (MMR). MMR preferentially protects early-replicating, gene-rich regions (Supek & Lehner 2015 Nature), lowering their mutation rates. Within such regions, the active-chromatin histone mark H3K36me3 additionally associates with lower mutation rates (Supek & Lehner 2017 Cell).

The enemy within. Mutations do not always result from copying mistakes during cell division, nor from external damage to DNA – endogenous factors can sometimes corrupt the genetic message. By cataloguing clustered mutations, we suggest that the error-prone DNA polymerase eta (POLH protein) generates many (≥40%) mutations at A:T nucleotide pairs in several human tissues (Supek & Lehner 2017 Cell). Furthermore, we find that the APOBEC3A antiviral defence system commonly generates a clustered 'mutation fog' (omikli) in human cancers (Mas-Ponte & Supek 2021 Nat Genet). Both the POLH and the APOBEC3A mutation processes are very likely to generate impactful mutations, because they are (unfortunately) enriched in gene-rich domains in the genome, likely because they can piggyback onto various DNA repair processes. Thus under certain conditions DNA repair can be corrupted, accelerating somatic evolution.